Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models741Deals30d269Policy30d65, 14 pendingSources1,014

Gemini 2.5 Pro

GoogleReleased Jun 5, 2025Current

Score basis

Gemini 2.5 Pro is capable in multimodal tasks and factuality; and behind the leaders in long context, reasoning, coding, instruction following, and math. Too few results yet to rate agentic tasks, safety, or multilingual tasks.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Capable2 capabilities
  1. Multimodal−19.0%5 of 6
  2. Factuality−21.9%4 of 4
Limited5 capabilities
  1. Long Context−26.2%2 of 3
  2. Reasoning−34.8%5 of 6
  3. Coding−38.3%5 of 10
  4. Instruction Following−40.9%2 of 3
  5. Math−56.2%2 of 5
Not rated3 capabilities

Too few results yet to rate Agentic, Safety or Multilingual.

Price

$1.25input$10.00outputper million tokens

From Artificial Analysis · 4 providers tracked · All prices

Costs more than 78% of 330 priced models · 3:1 input-to-output blend, log scale

Evidence

163results on92benchmarks

  • 66 independently verified
  • 28 aggregator
  • 24 vendor-reported
  • 45 cross-referenced

From 38 sources · latest Oct 8, 2026 · How verification works

API features

Tool callingStructured outputsReasoning

As listed by OpenRouter

Research

53 papers reference Gemini 2.5 Pro

Gemini 2.5 Pro benchmark results

163 results on 92 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Multimodal

CapableFull multimodal ranking

19.0% behind the leader5 of 6 ranked benchmarks measured

Show 6 more multimodal resultsHide 6 multimodal results

Factuality

CapableFull factuality ranking

21.9% behind the leader4 of 4 ranked benchmarks measured

Long Context

LimitedFull long context ranking

26.2% behind the leader2 of 3 ranked benchmarks measured

Show 1 more long context resultHide 1 long context result

Reasoning

LimitedFull reasoning ranking

34.8% behind the leader5 of 6 ranked benchmarks measured

Show 18 more reasoning resultsHide 18 reasoning results

Coding

LimitedFull coding ranking

38.3% behind the leader5 of 10 ranked benchmarks measured

Show 10 more coding resultsHide 10 coding results

Instruction Following

LimitedFull instruction following ranking

40.9% behind the leader2 of 3 ranked benchmarks measured

Show 3 more instruction following resultsHide 3 instruction following results

Math

LimitedFull math ranking

56.2% behind the leader2 of 5 ranked benchmarks measured

Agentic

Not enough dataFull agentic ranking

0 of 7 ranked benchmarks measured

Show 1 more agentic resultHide 1 agentic result

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 90 more resultsHide 90 results

Gemini 2.5 Pro: common questions

Who makes Gemini 2.5 Pro?

Gemini 2.5 Pro is made by Google.

When was Gemini 2.5 Pro released?

Gemini 2.5 Pro was released on Jun 5, 2025, according to Artificial Analysis.

What is Gemini 2.5 Pro good at?

Gemini 2.5 Pro is capable in multimodal tasks and factuality; and behind the leaders in long context, reasoning, coding, instruction following, and math. Too few results yet to rate agentic tasks, safety, or multilingual tasks.

How much does Gemini 2.5 Pro cost?

Gemini 2.5 Pro costs $1.25 per million input tokens and $10.00 per million output tokens, according to Artificial Analysis. We track its price at 4 providers. At a mix of three input tokens to one output token, it costs more than 78% of the 330 priced models we track.

How many benchmarks has Gemini 2.5 Pro been tested on?

We track 163 results for Gemini 2.5 Pro on 92 benchmarks from 38 sources, 66 of them independently verified. The latest was recorded on Oct 8, 2026.

Which API features does Gemini 2.5 Pro support?

OpenRouter lists tool calling, structured outputs, and reasoning for Gemini 2.5 Pro.

About this record

Where Gemini 2.5 Pro's numbers come from, and every name it appears under.

Tracked since
Apr 27, 2026
Newest source mention
Sep 10, 2026

Where the results come from

Verification: 163 scores · 66 independently verified · 28 aggregator-attributed · 45 vendor cross-reference · 24 vendor-reported. How these tiers are assigned

From 38 sources on 16 sites. Hugging Face supplies 48 of them; the 66 independently verified results come from 15 sites. Bars are coloured by trust tier.

  • huggingface.co48
  • artificialanalysis.ai30
  • api.llm-stats.com21
  • livecodebench.github.io12
  • matharena.ai11
  • arcprize.org10
  • storage.googleapis.com9
  • epoch.ai4
  • raw.githubusercontent.com4
  • aider.chat3
  • 99franklin.github.io2
  • datasets-server.huggingface.co2
  • lmarena.ai2
  • simple-bench.com2
  • swebench.com2
  • labs.scale.com1

Also known as

gemini 2.5 pro (03-25 preview)gemini 2.5 pro (03-25)gemini 2.5 pro (05-06)gemini 2.5 pro (06-05)gemini 2.5 pro (2025-05-06)gemini 2.5 pro (agent)gemini 2.5 pro (best-of-32)gemini 2.5 pro (jun 2025)gemini 2.5 pro (preview, thinking 1k)gemini 2.5 pro (preview)gemini 2.5 pro (thinking 16k)gemini 2.5 pro (thinking 1k)gemini 2.5 pro (thinking 32k)gemini 2.5 pro (thinking 8k)gemini 2.5 pro preview (jun 2025)gemini 2.5 pro preview (mar' 25)and 16 more